Chronic kidney disease is a worldwide public health problem, with increasing incidence and prevalence, poor outcomes and high cost. The overall aim of this research application is to use individual patient data meta-analysis to provide the evidence necessary to improve clinical care for patients with chronic kidney disease. The current grant period focused on pooling data from randomized clinical trials of angiotensin-converting-enzyme inhibitors to slow the progression of non-diabetic kidney disease. The conduct of this research has resulted in high impact original articles, incorporation of the findings into reviews and clinical practice guidelines, and creation of an infrastructure for stimulating collaborative research on other topics related to chronic kidney disease. In the next grant period will apply individual patient data meta-analysis to another problem central to the diagnosis and management of patients with chronic kidney disease - estimation of kidney function (glomerular filtration rate, GFR) from serum measurements. The specific aims are 1) Establish a collaboration of investigators from a wide range of studies in which GFR was measured using renal clearance of iothalamate and pool individual patient data into a single database. We have already secured collaborations to pool data on >6,000 patients (>31,000 GFR measurements) from research studies and > 14,000 patients (>23,000 GFR measurements) from clinical populations. 2) Develop and validate prediction equations using serum creatinine for estimating GFR. Using state of the art statistical and laboratory methods and diverse patient populations, we will be able to rigorously evaluate the validity of these equations in a wide range of clinical situations. 3) Assess cystatin as a filtration marker, and develop and validate prediction equations using serum cystatin for estimating GFR. 4) Use the new equations to estimate the prevalence of decreased GFR levels in the US population. 5) Characterize precision of GFR estimates in individuals and incorporate these precision estimates when estimating prevalence of GFR levels in populations. The principal investigator and co-investigators have expertise in nephrology, clinical chemistry, biostatistics, and epidemiology and have extensive experience with interpretation of GFR measurements, laboratory methods, and statistical techniques necessary to complete the project. The results will have a major impact on the care of patients with chronic kidney disease.